Classifications

• • A—HUMAN NECESSITIES
  • A22—BUTCHERING; MEAT TREATMENT; PROCESSING POULTRY OR FISH
  • A22B—SLAUGHTERING
  • A22B5/00—Accessories for use during or after slaughtering
  • A22B5/0064—Accessories for use during or after slaughtering for classifying or grading carcasses; for measuring back fat
  • A22B5/007—Non-invasive scanning of carcasses, e.g. using image recognition, tomography, X-rays, ultrasound
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
  • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
  • G01N33/02—Food
  • G01N33/12—Meat; Fish

Definitions

• the field of the present invention is prediction of meat palatability and yield. More specifically, the present invention relates to the prediction of meat palatability and yield by use of image analysis ( I A) to determine the color parameters L* (psychometric lightness), a* (red vs. green), and b* (yellow vs. blue) or the tissue density of the lean and fat portions of a meat animal carcass or cut.
• I A image analysis
• Marbling score of a carcass has been shown to generally correlate with subsequent cooked meat palatability across a wide range of marbling levels for beef, pork, and lamb. However, between carcasses with the same marbling level, there are substantial differences in palatability. Other factors of the carcass believed to predict palatability include maturity score, muscle pH, and muscle color; these factors may be more valuable in the prediction of palatability of chicken, turkey, and fish. Among those with expertise in carcass examination, e.g. meat scientists and U.S. Department of Agriculture (USDA) graders, some of these factors can be scored and palatability predicted by assigning a USDA Quality Grade, given sufficient examination time.
• USDA U.S. Department of Agriculture
• Yield Grades are intended to estimate the cutability and composition of a carcass.
• Factors used to determine Yield Grades include hot carcass weight, ribeye area (cross-sectional area of the longissimus m. at the 12- 13th rib interface), estimated kidney, pelvic, and heart fat percentage, and actual and adjusted subcutaneous fat thickness at the carcass exterior.
• the time constraints described above for the calculation of Quality Grades also apply to the calculation of Yield Grades.
• the parameters that underlie the assignment of Quality Grades and Yield Grades are published by the USDA Agricultural Marketing Service, Livestock and Seed Division, e.g., for beef, the United States Standards for Grades of Carcass Beef.
• a device for scoring factors predictive of palatability of a meat carcass or cut in addition to an examination of the carcass or cut by a USDA grader would allow meat palatability to be more accurately predicted and USDA Quality Grades to be more accurately assigned. This would allow greater consumer confidence in the Quality Grading system, as well as any additional system for certification of conformance to product quality specifications, as would be desired in a "brand-name" program. In either event, more precise sortation of carcasses for determining meat prices would be allowed. This superior sortation would provide economic benefit to those at all segments of the meat production system: restaurateurs, foodservice operators, and retailers; packers; feed lot operators; and ranchers, farmers, and harvesters of pork, lamb, beef and dairy cattle, chicken, turkey, and various fish species. This superior sortation would also benefit scientists in the collection of carcass and cut data for research, and the previous owners of livestock in making genetic and other management decisions.
• One such device uses a "duo-scan" or “dual-component” image analysis system. Two cameras are used; a first camera on the slaughter floor scans an entire carcass, and a second camera scans the ribeye after the carcass is chilled and ribbed for quartering.
• video data are recorded from a beef carcass and transferred to a computer.
• a program run by the computer determines the percentages of the carcass comprised of fat and lean from the recorded image and additional data available, e.g. hot carcass weight.
• the quantities of cuts at various levels of lean that can be derived from the carcass are then predicted.
• the system is not able to predict palatability of the observed carcass for augmenting the assignment of a USDA Quality Grade or other purpose related to sorting carcasses based on eating quality.
• Such an apparatus It is desirable for such an apparatus to collect and process data and provide output within the time frame that a carcass is examined by a USDA grader under typical conditions in the packing house, commonly 5-15 sec. It is desirable for such an apparatus to return scores for at least one of, for example, color and color variability of lean tissue, color and color variability of fat tissue, extent of marbling, average number and variance of marbling flecks per unit area, average size of marbling and the variance of average marbling size, average texture, firmness of lean tissue, lean tissue density, fat tissue density and connective tissue density. It is desirable for the apparatus to use these measures to assign a grade or a score to carcasses in order that the carcasses can be sorted into groups that reflect accurate differences in cooked meat palatability. It is also desirable for the apparatus to use these measures to identify defect conditions in the meat such as, but not limited to, bruising, dark cutter or heat ring.
• an apparatus for measuring the cross-sectional surface area of an exposed, cut muscle e.g. ribeye
• the apparatus uses this measure to assign a grade or score to carcasses in order that the carcasses can be sorted into groups that reflect accurate differences in yield.
• this apparatus also measure relative areas of cross-section surfaces comprised of fat and/or bone.
• an apparatus for measuring, predicting, and sorting carcasses on the bases of any combinations of palatability, defect conditions, and yield Further, it is desirable for such an apparatus to be portable, e.g. small and lightweight.
• the present invention is related to a method for predicting the palatability of meat, comprising: providing image data related to at least a portion of the meat; analyzing the image data to distinguish at least one area of interest of the meat; analyzing the image data corresponding to each area of interest to measure at least one characteristic of the area of interest based on the image data; predicting the palatability of the meat based on the characteristic.
• the present invention is also related to an apparatus for predicting the palatability of meat, comprising: an imaging device adapted to provide an image data of at least a portion of the meat; a data processing unit adapted to execute program instructions; a program storage device encoded with program instructions that, when executed, perform a method for predicting the palatability of meat, the method comprising: analyzing the image data to distinguish at least one area of interest of the meat; analyzing the image data corresponding to the area of interest to measure at least one characteristic of the lean section based on the image data; and predicting the palatability of the meat based on the characteristic.
• FIG. 1 shows a schematic view of an apparatus of the present invention.
• FIG. 2 shows a flowchart of a method of the present invention.
• FIG. 3 shows a flowchart of a computer program analyzing image data to distinguish at least one area of interest of the meat, analyzing the image data corresponding to the area of interest to measure at least one characteristic of the area of interest based on the image data.
• the present invention provides an image analysis (IA) system for scoring factors predictive of the palatability of a meat animal carcass.
• the IA system may be any type of imaging system known to those of skill in the art, such as a camera, tomographic imaging, magnetic resonance imaging (MRI), sound wave imaging, radio wave imaging, microwave imaging, or particle beam imaging, and is preferably a color video IA system.
• the I A system includes an imaging device 12, preferably a 3- CCD color video camera, preferably mounted in an enclosurell.
• the imaging device 12 optionally features an illumination system 26 mounted either on the imaging device, on the imaging device enclosure, or not on the imaging device but in the imaging device enclosure.
• the illumination system 26 may be any light emitting device known to those of skill in the art, or a source of energy of equivalent function meant to impinge on the sample of meat for measurement in the parts of the energy spectrum corresponding to the sensitivity ranges required by the type of imaging device 12, including visible light, infrared light, ultraviolet light, x-rays, gamma rays, electrons, positrons, electrical fields, magnetic fields, sonic wave, ultrasonic wave, infrasonic wave or microwaves.
• the IA system also includes a data processing unit 16, the data processing unit 16 interfaced with a program storage device 20 by a program storage device interface 18, and at least one output device 24 by an output device interface 22.
• the program storage device 20 contains a computer program or programs required for proper processing of image data, preferably color video image data, by the data processing unit 16.
• the data processing unit 16 is linked to, and receives data from, the video camera 12 via either a transfer cable 14 or a wireless transmission device (not shown).
• the data processing unit 16 comprises a standard central processing unit (CPU), and, where necessary or appropriate, preferably also a software module or hardware device for conversion of analog data to digital data, and processes image data according to instructions encoded by a computer program stored in the program storage device 20.
• Image data can be used in subsequent calculation of the values of characteristics, the values being predictive of palatability, the characteristics including color and color variability of lean tissue, color and color variability of fat tissue, extent of marbling, average number and variance of marbling flecks per unit area, average size of marbling and the variance of average marbling size, average texture of marbling and lean tissue, firmness of lean tissue, density of lean tissue, density of fat tissue, and density of connective tissue.
• These values can then be used to sort meat (herein defined as a meat animal carcass, side, or cut, or any portion of a carcass, side, or cut) into groups that vary in predicted subsequent cooked eating quality.
• the color parameters L*, a*, and b*, or the tissue density parameters can also be used to calculate the values of factors predictive of yield, such as the cross-sectional area of a muscle of interest and other surrounding organs such as fat, bone, and connective tissue. These values can then be used to sort meat into groups that vary in predicted composition.
• the color parameters L*, a*, and b* can also be used to calculate the values of factors predictive of defect conditions of the meat, such as bruising, dark cutter and heat ring. These values can then be used to denote the carcass as defective or to adjust or otherwise alter the quality sortation decision.
• the data processing unit 16 is linked to, and transmits results of data processing to, at least one output device 24 by output device interface 22.
• results of data processing can also be written to a file in the program storage device 20 via program storage device interface 18.
• An output device 24 can be a video screen, printer, or other device. It is preferred that at least one output device 24 provide a physical or electronic tag to label the meat 10 with results of data processing, in order to facilitate sortation of meat animal carcasses, cuts, or both into groups with similar predicted palatability and/or yield.
• the present invention also provides a method of predicting the palatability of the meat 10 and determining the cross-sectional area of the meat 10.
• image data collected from meat 10 is recorded by the imaging device 12, processed by the data processing unit 16, and the values of palatability and/or muscle cross-sectional area is output by the output device 24 to augment the observations made by a USDA line grader, or other operator responsible for sorting or characterizing meat animal carcasses, in order to allow more accurate assignment of Quality Grades, Yield Grades, defect conditions, and/or other sorting or classification criteria based on the characteristics.
• An apparatus for use in the present invention comprises an imaging device 12 and a data processing unit 16.
• the imaging device 12 can be any such imaging device known to those of skill in the art, such as a camera, tomographic imaging (i.e. CAT, PET) device, magnetic resonance imaging (MRI) device, sound wave imaging device, radio wave imaging device, microwave imaging device, or particle beam imaging device. If the imaging device is a camera, the camera is at least one of a photographic camera, a digital still camera, and a video camera.
• the camera responds to light in at least one portion of the light spectrum, each such portion consisting of a band of, such as a segment of ultraviolet wavelengths (200 nm to 400 nm), visible wavelengths (400nm to 700nm), infrared wavelengths (700 nm to 10 m), or po r tions thereof.
• the image can be obtained by at least one of x-ray tomography, and particle beam tomography, such as computer axial tomography (CAT) or positron emission tomography (PET).
• CAT computer axial tomography
• PET positron emission tomography
• the image is a sound wave image
• it can be obtained by ultrasound, B-mode ultrasound, or infrasonic imaging.
• These devices produce non-invasive cross-sectional and three- dimensional images resulting from the transmission and reflection characteristics of the specimen to the frequency of sound wave applied, where intensity is a function of object cross- sectional density.
• the imaging device 12 It is important for the imaging device 12 to provide output within the time frame allotted for meat carcass examination, typically 5-15 seconds. Preferably the output is in real-time.
• Such real-time output can be the same technology as the viewfinder on a known camcorder or video camera, the real-time output can be the same technology as a known digital camcorder, the real- time output can be a known computer-generated real-time display as are known in videoconferencing applications, or can be any other technology known to those of skill in the art.
• the imaging device 12 It is preferable for the imaging device 12 to be a color video camera, for reasons discussed below. It is also preferred that the imaging device 12 be small and lightweight, to provide the advantages of portability and flexibility of positioning, i.e.
• the imaging device 12 be durable, in order to better withstand the environment of the packing plant.
• the power source of the imaging device 12 can be either direct current, i.e. a battery secured to electrical contacts from which the imaging device 12 can draw power, or alternating current provided from either an electrical outlet or from the data processing unit 16.
• An illumination system 26 can optionally be used to illuminate the meat with energy in the useful spectrum of the imaging device. This is desirable when using visible light imaging and the ambient lighting is poor or uneven or when it is desired to examine regions of the meat 10 that are not illuminated by ambient light, or when the spectral sensitivity of the imaging device is not in the visible light part of the electromagnetic spectrum.
• Any known or future developed illumination system 26 can be used, e.g. a lamp (incandescent, fluorescent, etc.), a laser, etc. for visible and near- visible portions of the spectrum , x-rays, gamma rays, electrons, electrical fields, magnetic fields, sonic beam, ultrasonic beam, infrasonic beam,or microwaves.
• the power source 42 of the illumination system 26 can be either direct current, i.e. a battery 45, or alternating current drawn from either an electrical outlet 50, the imaging device 12, or the data processing unit 16. It is preferred that the illumination system 26 be small and lightweight, for reasons discussed in reference to the imaging device 12, above.
• the illumination system 26 can be mounted on the imaging device 12, on the outer surface of an imaging device enclosure 11, or within an imaging device enclosure 11, which is described hereafter.
• the imaging device 12 and optional illumination system 26 can be unenclosed or enclosed.
• the imaging device 12 is enclosed in an imaging device enclosure 11 for protection against the environment of packing and processing plants. It is important for the imaging device enclosure 11 to provide a first aperture 13 for the lens of the imaging device 12 to observe the meat 10.
• the illumination system 26 can be mounted either on the outer surface of the imaging device enclosure 11 or within the imaging device enclosure 11. If mounted within the imaging device enclosure 11, the illumination system 26 can be mounted on the imaging device 12. If the illumination system 26 is mounted in the imaging device enclosure 11, it is important for an aperture to be provided for illumination of the meat 10, either the first aperture 13 used by the lens of the imaging device 12 or a second aperture. In either case, the aperture can be unencased or it can be encased by a pane of a transparent material 14, wherein "transparent" is defined as allowing the passage of substantially all of the energy type and wavelength emitted by the illumination system 26 or detectable by the imaging device 12.
• image data is to be transferred from the imaging device 12 to the data processing unit 16 by a transfer cable 44 connected therebetween, it is also important for the imaging device enclosure 11 to provide an aperture 31 for the cable to exit the enclosure.
• This aperture can be the first aperture 13 used by the lens of the imaging device 12, the second aperture that can be used by the illumination system 26, or a third aperture 31.
• the imaging device enclosure 11 be constructed from a lightweight material and be only large enough to conveniently fit the imaging device 12, and optionally the illumination system 26 described above.
• alternating current is to be used as the power source of the imaging device 12, it is important for an aperture to be provided to pass the power cable from the imaging device 12 to the power source. Any one of the first, second, or third apertures can be used, or a fourth aperture can be used. If the aperture to be used is encased by a pane of transparent material, it is important to provide a second cable-passage aperture in the pane for passage of the power cable. Alternatively, both the power cable and the data-transfer cable can exit the imaging device enclosure through a single cable-passage aperture.
• the imaging device enclosure can be designed with features to more readily allow user grip and manipulation, e.g.
• wall, ceiling, or tripod mounting can be to motorized rotatable heads for adjusting imaging device angle and focal length.
• the imaging device enclosure can be designed to be easily opened to allow for convenient maintenance of the imaging device 12 or replacement of a battery if direct current is used as the power source of the imaging device 12.
• Maintenance of the illumination system 26 may also be needed, and preferably in this option will be allowed by the same easy-opening design described for the imaging device 12.
• the easy-opening design can be affected by the use of screws, clamps, or other means widely known in the art. Ease of maintenance is desirable to minimize any downtime that may be encountered.
• image data After image data is captured by the imaging device 12, it is transferred in real- time to the data processing unit 16.
• Data can be transferred by a transfer cable 14 or by a wireless data transmission device (not shown). In most situations, transfer cable 14 is the preferred medium of transmission based on superior shielding and lower cost. In situations where the imaging device 12 and data processing unit 16 are widely separated, a wireless data transmission device (not shown) can be a more practical medium of transmission. Any technique of data transfer presently known or developed in the future by those of skill in the art can be used.
• the video image data can be sent from the video camera 12 to the data processing unit 16 as either analog or digital data.
• a "data processing unit” 16 is defined as including, but not limited to, desktop computers, laptop computers, handheld computers, and dedicated electronic devices. Any data processing unit presently known or developed in the future by those of skill in the art can be used in the present invention. In one embodiment of the present invention, the data processing unit 16 can be small and lightweight to provide portability.
• the data processing unit 16 can be a microcomputer, minicomputer, or mainframe that is not portable.
• the present invention is not limited to any specific data processing unit, computer, or operating system.
• An exemplary embodiment, but one not to be construed as limiting, is a PC- compatible computer running an operating system such as DOS, Windows, or UNIX.
• the choice of hardware device or software module for conversion of analog data to digital data for use in the present invention is dependent on the imaging device 12, data processing unit 16, and operating system used, but given these constraints the choice will be readily made by one of skill in the art.
• the imaging device 12 is a visible light sensitive device such as a color camera
• the data processing unit 16 comprises a software module that converts RGB color to L*a*b* color.
• An exemplary software module for this purpose is sold in HunterLab Color Vision Systems (Hunter Associates Laboratory, Inc.).
• the data processing unit 16 include other input devices, e.g. a keyboard, a mouse or trackball, a lightpen, a touchscreen, a stylus, a bar code reader, etc., to allow convenient exercise of user options in camera and software operation, data processing, data storage, program output, etc.
• input devices e.g. a keyboard, a mouse or trackball, a lightpen, a touchscreen, a stylus, a bar code reader, etc.
• a program storage device 20 examples of program storage devices being a hard drive, a floppy disk drive, a tape drive, a ROM, and a CD-ROM, among others
• An exemplary code for such a program or programs is given in an appendix hereto.
• An exemplary flowchart for such a program or programs is given as FIG. 3.
• the image data can be analyzed for color scale parameters. If it is desired to conform to international standard, the image data can be analyzed for the color scale parameters L*, a*, and b*, as defined by the Commission Internationale d'Eclairage (CIE). A set of L*a*b* parameters is recorded for each frame. L*, a*, and b* are dimensions of a three-dimensional color space which is standardized to reflect how color is perceived by humans.
• CIE Commission Internationale d'Eclairage
• the L* dimension corresponds to lightness (a value of zero being black, a value of 100 being white), the a* dimension corresponds to relative levels of green and red (a negative value being green, a positive value being red), and the b* dimension corresponds to relative levels of blue and yellow (a negative value being blue, a positive value being yellow).
• the system can capture pixelated images from areas of 12 to 432 square inches (75 to 2700 cm ) from the muscle of interest, comprising up to 350,000 pixels per measurement, and determine L*, a*, and b* for sub-regions of the image frame comprising at least one pixel each. In all embodiments, it is desirable for determination of L*a*b* to be performed using the HunterLab software conversion or similar module.
• L*a*b* At least one of the L*, a*, and b* components can be used in subsequent data processing.
• the image data can be further analyzed to identify areas of interest corresponding to factors predictive of palatability using any such methods of analysis known to those of skill in the art, including, among others, image segmentation, histogram thresholding, spatial analysis, pattern matching, pattern analysis, neural network, region growing, and focus of attention methods, as described in numerous references, such as The Image Processing Handbook 3 rd Edition, 1999, John C. Russ, CRC Press.
• a program then calculates several parameters of the image for each frame.
• the program outlines the muscle of interest by choosing areas that have tolerances of b* compatible with muscle.
• Areas with values of b* compatible with the muscle of interest are then examined for their L* and a* scores for verification and rejection of surrounding tissues invading the outline of the muscle of interest. Further examination need not be performed on areas with L*, a*, and b* scores suggestive of bone, connective tissue, and fat.
• the surface area of the cross-section of the muscle of interest is determined.
• the lean tissue and fat tissue of the muscle can be distinguished and raw L*, a*, and b* scores, and variation in L*, a* and b* scores across the area of interest, for the lean tissues of the muscle can be determined.
• These scores can then be sent to the output device 24 to be displayed in numerical format and/or retained to calculate quality- and yield-determining characteristics as described below. It is known that, among other characteristics, higher values of b* for lean tissues of muscle correlate with greater tenderness (Wulf et. al., 1996).
• the fat color and color variability of intermuscular fat can also be determined.
• determinations can be made of the quantity, distribution, dispersion, texture, and firmness of marbling (intramuscular fat deposited within the muscle).
• the quantity of marbling can be determined by calculating the percentage of muscle surface area with L*, a*, and b* scores compatible with fat tissue.
• the distribution and dispersion of marbling can be determined.
• the portion of the image derived from the muscle of interest can be divided into subcells of equal size. A size of 64 x 48 pixels can be used.
• the number of marbling flecks can be determined as the number of discrete regions with L*, a*, and b* values corresponding to fat, and the average number of marbling flecks per subcell can be calculated. The variance of numbers of marbling flecks across all subcells can be calculated as well.
• the average size of each marbling fleck can be determined throughout the muscle of interest from the number of pixels within each discrete region with L*, a*, and b* values corresponding to fat.
• the variance of marbling size across all marbling flecks can be calculated as well.
• the texture and fineness of marbling can also be measured. It is well known that generally, greater amounts of more uniformly distributed and finer-textured marbling reflect a higher marbling score and thus meat of higher eating quality.
• the program can use L*, a*, and b* data to calculate the average texture, i.e. cross- sectional surface roughness, of the muscle, and also the firmness of the lean tissue of the cross- sectional muscle. It is well known that the surface roughness of a muscle is inversely correlated with tenderness, and greater firmness is correlated with flavorfulness.
• the program can determine the density of the lean tissue, the density of the fat tissue and the density of the connective tissue.
• characteristics of the area of interest of the meat 10 include, but are not limited to, the color of the lean tissue, the color variation of the lean tissue, the color of fat tissue, the color variation of the fat tissue, a marbling quantity, a marbling distribution, a marbling dispersion, a marbling texture, a marbling fineness, an average texture of the lean tissue, a firmness of the lean tissue, a surface area of the lean section, the density of the lean tissue, the density of the fat tissue and the density of the connective tissue.
• Quantities of the non-lean section of the meat including but not limited to the color of fat, the density of the lean tissue, the density of the fat tissue, the density of the connective tissue and the relative areas of cross-section surfaces comprised of fat, bone, and/or connective tissue, may be calculated as well.
• Other characteristics that one of skill in the art of meat science can readily see may be calculated from the values of L*, a*, and b* and that can be predictive of palatability can be calculated by the program, and any such characteristics are considered to be within the scope of the present invention.
• the program can output to the output device 24 calculated values of any or all of the characteristics given above: color of lean tissue, color variation of lean tissue, color of fat tissue, color variation of fat tissue, extent of marbling, average number of marbling flecks per unit area, variance of marbling flecks per unit area, average size of marbling, variance of the average size of marbling, texture and fineness of marbling, average texture of lean tissue, firmness of lean tissue, the density of the lean tissue, the density of the fat tissue and the density of the connective tissue.
• the calculated values of the characteristics, if output are displayed as alphanumeric characters that can be conveniently read by the operator.
• other parameters not determinable by means of analyzing the image data may be used to augment the characteristics determined from the image data, including, but not limited to, pH, connective tissue quantity, connective tissue solubility, sarcomere length, protease enzymatic activity, calcium measure, electrical impedance, electrical conductivity, and tissue density.
• These additional characteristics of the meat may be used singly or in combination to improve the accuracy of prediction of palatability over that obtained solely using characteristics measured from the image data.
• the area of the cross-sectional surface of the muscle portion of the meat 10 can be calculated and output to the output device 24.
• image data may be used to augment the characteristics determined from the image data, including, but not limited to, subcutaneous fat depth.
• additional characteristics of the meat may be used singly or in combination to improve the accuracy of prediction of Yield Grades over that obtained solely using characteristics measured from the image data.
• results reported by the program can be output to any output device 24, such as a screen, printer, speaker, etc. If operator evaluation of the results is desired, results can preferably be displayed on a screen. Preferably, the screen is readily visible to the grader, evaluator, or operator at his or her stand. Alternatively, or in addition, it is preferable that results be printed or output in such a manner that the outputted- results can be transferred and affixed to the meat 10.
• the manner for outputting results can be text, symbols, or icons readable by personnel either in the packing plant or at later points in the meat production system.
• the manner for outputting results can be a barcode or other object that can be read by appropriate equipment and decoded into forms readable by personnel at various points in the production system.
• Output results can be affixed to the meat 10 by methods well-known to the art, which include, but are not limited to, pins, tacks, and adhesive.
• the power source 50 of the data processing unit 16 can be either direct current, i.e. a battery, or alternating current drawn from an electrical outlet.
• the data processing unit 16 can be mounted in a data processing unit enclosure 15 or in the imaging device enclosure 11, or can be unenclosed.
• the data processing unit 16 is a microcomputer, minicomputer, or mainframe computing resource present in the plant or facility where the apparatus is used, enclosure is not required.
• the data processing unit 16 is a separate, stand-alone, portable entity, preferably the data processing unit 16 is mounted in a data processing unit enclosure 15.
• the data processing unit enclosure It is important for the data processing unit enclosure to provide an aperture 47 or apertures for output of data to or display of data by the output device 24. For example, if display is to be performed using a video screen congruent with the data processing unit 16, it is important for the data processing unit enclosure to provide an aperture 47 for observation of the video screen therethrough. Such an aperture can be unencased or it can be encased by a pane of transparent material 48, such as glass, plastic, etc. If display is to be performed by an external device, e.g. a remote monitor or printer, it is important for the data processing unit enclosure to provide an aperture (not shown) for passage of an output cable 22 therethrough.
• an external device e.g. a remote monitor or printer
• the data processing unit 16 is powered by alternating current, it is important for the data processing unit enclosure 15 to provide an aperture 49 for passage of a power cable therethrough. If it is desired to store outputs to an internal floppy disk drive (not shown), it is important for the data processing unit enclosure 15 to provide an aperture (not shown) for insertion and removal of floppy disks into and from the internal floppy disk drive therethrough. If it is desired to store outputs to an external program storage device 20, it is important for data processing unit enclosure 15 to provide an aperture 19 for passage of a data-transfer cable 18 therethrough.
• the data processing unit enclosure 15 is only large enough to conveniently fit the data processing unit 16, and is lightweight.
• the data processing unit enclosure 15 can be designed with features to more readily allow user manipulation, e.g. handles.
• the data processing unit enclosure 15 be amenable to easy opening to allow for convenient maintenance of the data processing unit 16. The easy-opening design can be affected by means described for the camera enclosure supra.
• the apparatus described above can be used in methods for predicting the palatability, yield, and/or defect conditions of, or augmenting the assignment of USDA or other international grade standards to, meat animal carcasses or cuts, or for sorting for other purposes (e.g. brand names, product lines, etc.).
• the first step involves collecting image data from the meat 10 using the imaging device 12.
• the second step involves processing the image data using the data processing unit 16.
• the third step involves using the results of the processing step in reporting quality-determining characteristics that can be used to augment USDA graders in the assignment of USDA Quality Grades, in reporting the areas of cross-sectional muscle surfaces that can be used to augment USDA graders in the assignment of USDA Yield Grades or other international grade standards, in reporting meat defect conditions, or in sorting the meat 10 based on specific requirements of, for example, a brand-name or product line program.
• the grader or operator's limited time to analyze the meat 10 can be focused on examining parameters most readily examined by a person, providing the grader or operator with more data for each sample of meat 10 in the same amount of time, and allowing more accurate prediction of palatability and assignment of Quality Grade and Yield Grade than is currently possible.
• this method allows required computations to be performed more quickly and accurately than is currently possible.
• Example I Segregation of beef carcasses with very low probabilities of tenderness problems
• a population of 324 beef carcasses was examined in an effort to segregate a subpopulation of carcasses with very low probabilities ( ⁇ 0.0003) of having ribeye shear force values of 4.5 kg or greater and subsequent unacceptably tough-eating cuts.
• ⁇ 0.0003 ribeye shear force values of 4.5 kg or greater
• 200 were certified to meet the above standard for tenderness.
• 17 head were preselected for the tender subpopulation on the basis of expert-determined (beef scientist or USDA Grading Supervisor) marbling scores of Modest, Moderate, or Slightly Abundant, the three highest degrees of marbling in the United States Standards for Grades of Carcass Beef.
• tenderness values for each of the remaining 247 head were predicted using a multiple regression equation using CIE a* values for lean and fat, as well as machine measured marbling percentage squared.
• the multiple regression equation determined that 123 out of 247 carcasses were predicted to have a probability of being not tender of 0.0003. These 123 carcasses were then segregated with the 77 that had been preselected, and certified as being tender. The remaining carcasses had a normal probability of 0.117 of having shear force values in excess of 4.5 kg.

Landscapes

• Engineering & Computer Science (AREA)
• Life Sciences & Earth Sciences (AREA)
• Food Science & Technology (AREA)
• Health & Medical Sciences (AREA)
• Chemical & Material Sciences (AREA)
• Biochemistry (AREA)
• Physics & Mathematics (AREA)
• Analytical Chemistry (AREA)
• Medicinal Chemistry (AREA)
• General Health & Medical Sciences (AREA)
• General Physics & Mathematics (AREA)
• Immunology (AREA)
• Pathology (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Biophysics (AREA)
• Image Analysis (AREA)
• Investigating Or Analysing Materials By Optical Means (AREA)
• Image Processing (AREA)

Applications Claiming Priority (3)

Publications (1)

Family

ID=22272202

Family Applications (1)

Country Status (6)

Cited By (1)

Families Citing this family (5)

Family Cites Families (10)

• 1999
  • 1999-08-19 MX MXPA01008411A patent/MXPA01008411A/es unknown
  • 1999-08-19 AU AU56826/99A patent/AU5682699A/en not_active Abandoned
  • 1999-08-19 CA CA002363089A patent/CA2363089A1/en not_active Abandoned
  • 1999-08-19 WO PCT/US1999/019027 patent/WO2000049400A1/en not_active Application Discontinuation
  • 1999-08-19 EP EP99943799A patent/EP1161679A1/en not_active Withdrawn
  • 1999-08-19 BR BR9917141-4A patent/BR9917141A/pt not_active Application Discontinuation

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events